fisherA2Z

Introduction

FisherA2Z is a Fisher forecasting code initially developed by Husni Almoubayyed for his PhD thesis Chapter 4. Tianqing Zhang refurbished the code in 2024 and prepared Zhang et al in prep. "A2Z" stands for the initial of Almoubayyed and Zhang.

This repository includes code used for the Fisher Information matrix computation used to assess the impact of photo-z modeling errors on 3x2pt inferences.

Installation

First, initialize a conda environment by

conda create -n 'fisher_env' python=3.8
conda activate fisher_env

Clone the repository, and install from source by

pip install -e .

Then you can add your conda environment to your jupyterLab by

conda install -c anaconda ipykernel
python -m ipykernel install --user --name=fisher_env

Code

The majority of the code is in fisher.py, and the Fisher class therein.

The Fisher class takes a CCL cosmo object and 3 iterables of length 5 each to specify the photo-z error model in terms of biases, standard deviations, and outlier fractions.

Examples

To get the fisher matrix for a certain case, it is sufficient to run

from fisher import Fisher
f = Fisher(cosmo=ccl_cosmo)
f.process()

then the Fisher information matrix will be stored in f.fisher.